Feature

Cross-Document Validation

Compare the same person across several documents and surface mismatches.

What is Cross-Document Validation in Mahad OCR?

Available with limits. When multiple documents belong to one candidate, shared fields such as name, date of birth and document number are compared so a mismatch is caught before it reaches your records. Input: Two or more processed documents linked to the same candidate. Limitation: Comparison is candidate-scoped: there is no global "validate everything" view yet, so documents must be attached to a candidate first.

Limited

What it does

When multiple documents belong to one candidate, shared fields such as name, date of birth and document number are compared so a mismatch is caught before it reaches your records.

How it works

  1. 1Process each document normally
  2. 2Attach documents to the same candidate
  3. 3Request the cross-validation view for that candidate
  4. 4Review the mismatch list

Sample result

Illustrative response shape using synthetic data.

{ "candidate_id": "…", "mismatches": [ { "field": "date_of_birth", "values": ["1990-02-11", "1990-11-02"] } ] }

Supported inputs

Two or more processed documents linked to the same candidate.

Fields returned

  • matching fields
  • mismatched fields
  • per-field comparison result

API

Candidate-scoped endpoints under /v1/candidates

Security & privacy

Comparison happens inside your tenant only.

Current limitations

Comparison is candidate-scoped: there is no global "validate everything" view yet, so documents must be attached to a candidate first.

Try it on your own document

Run the live demo, or create an account and get an API key.